The JSON files in a Flickr data download
photo_<id>.json file for each photo. The number in the photo's file name is the same number as in the JSON's file name. That number is how you pair them.What you get when you request your Flickr data
On Flickr, select your profile photo, then Settings, and under Your Flickr Data select Request My Flickr Data. Flickr prepares the files (it says this takes from a few hours to a few weeks) and emails you when the download links are ready in your account settings. The links expire, so download them soon. Source: Flickr Help, "Downloading content from Flickr".
The download is a set of ZIP files of two kinds:
| ZIP | What is inside |
|---|---|
| Account data | Small JSON files: one photo_<id>.json per photo or video, an albums.json, and files about your account such as account_profile.json. |
Photos and videos (data-download-1.zip, data-download-2.zip, …) | Your original photo and video files. A large account gets many of these parts. |
Why the titles and albums are not in the photos
Flickr gives the files back the way you uploaded them. Its help pages say the downloaded photos only contain the EXIF data that existed when they were uploaded, and that "Any details added to your photos after upload will be compiled in a separate JSON file". The EXIF data FAQ says the same about a title, description or tags added after upload: they "will be in a separate JSON file".
So the date and camera settings that were in a photo when you uploaded it are still there. What is missing from the file is everything you typed on Flickr: the title, the description, the tags and the albums.
How the photo files are named
Flickr puts the photo ID in every file name. These are the shapes found in real exports:
| File name | Kind | Flickr ID |
|---|---|---|
img_0642_3051711960_o.jpg | Photo with a title: title, ID, _o (original size) | 3051711960 |
30410016513_b941ab820b_o.jpg | Photo with no title: ID, a 10-character code, _o | 30410016513 |
fionas-cheer_54052605627.mov | Video: title and ID, with no _o | 54052605627 |
The matching JSON files are photo_3051711960.json, photo_30410016513.json and photo_54052605627.json.
What is inside photo_<id>.json
| Field | Meaning |
|---|---|
id | The Flickr photo ID. |
name | The title. |
description | The description. It can contain HTML, such as links and line breaks. |
date_taken | The date taken, as 2017-08-02 19:44:10, with no time zone. |
date_imported | The date the photo was uploaded to Flickr. |
tags | A list of tags, each one as {"tag": "tennessee"}. |
albums | The albums the photo is in, each with an id and a title. |
geo | The location, when the photo has one. Latitude and longitude are written as whole numbers, in millionths of a degree: "35685790" means 35.685790°. |
photopage | The address of the photo's page on Flickr. |
original | A direct address of the original file. |
privacy, license, comments, count_views… | Who can see the photo, its license, its comments and its counters. |
albums.json holds the list of albums. Each album has an id, a title, a description and photos, the list of photo IDs in that album.
How to match a photo to its JSON by hand
- Unzip the account data ZIP and the photo ZIPs.
- Take the photo's file name and remove the extension and the final
_o. - Read the last number in what is left: that is the Flickr ID. For an untitled photo, the ID is the first part, before the 10-character code.
- Open
photo_<that number>.jsonin a text editor. The title is inname, the albums inalbums.
This works for a handful of photos. For thousands of photos, people use Python scripts with ExifTool, or a tool that does the whole match in one pass.
Let AlbumTidy do the matching
Drop the ZIPs from Flickr and get your photos back in album folders, with titles, descriptions, tags and missing dates written into each JPEG. Free for 25 photos, nothing uploaded.
Tidy my Flickr downloadWhat AlbumTidy does with these files
- Matches every photo and video with its
photo_<id>.jsonby Flickr ID, across all the ZIPs. - Writes the title, description and tags into each JPEG (XMP), the description into EXIF
ImageDescription, the date taken into EXIFDateTimeOriginaland the location into EXIF GPS, each one only when the photo does not have it yet. - Puts each photo in a folder named after its first album and renames it
YYYY-MM-DD_Title.jpg. - Writes
manifest.csv(ID, original name, new name, title, description, date taken, date uploaded, tags, albums, GPS, Flickr URL) andunmatched.csv(photos with no JSON, JSON files with no photo). - Leaves videos, PNG, GIF and HEIC files byte for byte as they are, and never modifies the ZIPs you downloaded.